Representativeness of the “Fiesole Misurata”study database for use in pharmaco-epidemiological investigations on adherence to antihypertensive medications Running head: Representativeness of the “Fiesole Misurata” study database
Francesco Lapi 1,2,3, Ersilia Lucenteforte 1,*, Martina Moschini 1, Roberto Bonaiuti 1, Marina Di Pirro 1, Alessandro Barchielli 4, Silvia Benemei 1, Maddalena Belladonna 5, Nicola Nesti 5, Raffaele Coppini 1, Margherita Taras 6, Alfredo Vannacci 1, Andrea Ungar 5, Alessandro Mugelli 1.
1. Department of Preclinical and Clinical Pharmacology, Centre for Molecular Medicine (CIMMBA), University of Florence, Italy
2. Centre for Clinical Epidemiology and Community Studies, Sir Mortimer B. Davis Jewish General Hospital, Montreal, Quebec, Canada
3. Department of Epidemiology, Biostatistics, and Occupational Health, McGill University, Montreal Quebec, Canada
4. Department of Epidemiology, Local Health Authority n°10, Florence, Italy 5. Unit of Gerontology and Geriatrics, Department of Critical Care Medicine and
Surgery, University of Florence and Azienda Ospedaliero-Universitaria Careggi, Florence, Italy
6. Fiesole Municipality, Fiesole (Florence), Italy
Keywords: “Fiesole Misurata” ; database; representativeness; adherence; antihypertensive medications.
Correspondence to: Ersilia Lucenteforte, ScD, PhD
Department of Preclinical and Clinical Pharmacology - Centre for Molecular Medicine (CIMMBA)
University of Florence
viale G. Pieraccini 6 - 50139 Florence, Italy tel. 055 4271333; fax 055 4271280
ABSTRACT
Background and Aims: Poor adherence to medications is a major health concern especially among older subjects. To plan future studies to improve adherence, an epidemiological study, called “Fiesole Misurata”, was conducted. The aim of the present paper was to verify the representativeness of the database in evaluating the AntiHyperTensives (AHTs)-taking behaviour.
Methods: Demographic records of all subjects aged ≥ 65 years (n=2,228) living in the community of Fiesole (Florence, Italy) was retrieved from the Registry Office of Fiesole Municipality. The corresponding healthcare records were obtained from administrative archives of the Local Health Authority (claim dataset). Moreover, a cohort of subjects aged ≥65 years (n=385) living in the community was screened by means of a multidimensional geriatric evaluation (cross-sectional dataset).
Results: In claim dataset, biyearly prevalences of hospitalization for ischemic
cardiomyopathy, heart failure, and stroke were 3.7%, 3.0%, and 3.2%, respectively. In the cross-sectional dataset, prevalences were 11.2%, 6.7%, and 7.1%, respectively. The most used drugs were angiotensin-converting enzyme (ACE) inhibitors (43.6% in the claim dataset, 45.3% in the cross-sectional dataset) and diuretics (35.6% and 47.0%, respectively). Among the incident users of AHTs, 63.5% was highly adherent (≥80%) over the first six months of follow-up, while 14.3% and 22.2% were intermediate (40-79%) and low (<40%) adherent. The percentage of high adherers decreased with time and reached 31.2% at the 24th month.
Conclusions: These findings indicate that “Fiesole Misurata” study database can be used to develop future strategies aimed at improving the adherence to AHTs in older individuals.
INTRODUCTION 1
Poor adherence to medications is a major health concern [1] especially among older 2
subjects. Generally, when all drug categories are taken into account, the proportion of 3
non-adherent older subjects varies from 40 to 75% [2]. This issue is particularly 4
relevant for chronic asymptomatic diseases, such as hypertension, dyslipidaemias, 5
diabetes,or other age-related disorders. 6
In specific, most of the fatal CardioVascular (CV) events occur in individuals 7
aged 65 or older, in which the prevalence of hypertension is greater than in younger 8
adults and leads to half and approximately to two-thirds of Coronary Heart Diseases 9
(CHD), and cerebrovascular events, respectively [3-6]. Therefore, an inadequate Blood 10
Pressure (BP) control could significantly increase the risk of death because of ischemic 11
heart disease and stroke [7-9]. 12
Although data on the clinical burden of non-adherence to AntiHyperTensives 13
(AHTs) among older individuals are scanty, prior findings raised concerns about the 14
relevance of non-adherence to AHTs, that hampers the effectiveness of these 15
medications. Specifically, it has been demonstrated that among middle-aged patients an 16
high adherence to AHTs is associated with a significant decreased risk (38%) of major 17
CV events when compared with a low adherence [10]. 18
The basis of poor medication-taking behaviour is multifactorial, as demonstrated 19
by the strict relationship between a greater therapeutic complexity and a low adherence 20
to CV medications [11]. In this context, the older community-dwelling people are the 21
best example of therapeutic complexity, given the higher number of coexistent diseases 22
and concomitant medications as well as the co-occurrence of other conditions, such as 23
functional and cognitive impairments, age-related physiological complications (i.e., 24
reduced liver and kidney function), which cannot be necessarily ascribed to a specific 1
organic disease [12]. 2
There are many unanswered questions on the most effective strategies for 3
improving medications adherence in older subjects. They can be addressed with the use 4
of electronic healthcare databases [13]. Claim repositories, which comprise all 5
reimbursed drug prescriptions, hospital admissions diagnoses, and mortality registers 6
can be valid tools in implementing intervention strategies. Nevertheless, claim 7
databases are not designed for a specific research question, so certain variables (i.e. 8
values of BP, disability and cognitive status) are often unavailable [2, 14]. For this 9
reason, research on antihypertensive non-adherence in the elderly, cannot be 10
exhaustively satisfied with the use of claim database since some confounders are not 11
measurable. 12
To overcome this issue and with the aim to plan future studies to improve 13
adherence, an epidemiological study, called the “Fiesole Misurata” study, was 14
conducted in Fiesole, a small town of Tuscany, Italy, located in the hill north of 15
Florence, and an ad hoc database was assembled. The name of the study can be 16
translated as “Measuring Fiesole” since the database comprises several “measurements” 17
(overall representing a multidimensional evaluation) of the population living in Fiesole, 18
including socio-demographic and clinical information of all older (≥65 years) residents, 19
who were retrospectively collected using claims data. In addition, a cohort of subjects 20
underwent a multidimensional geriatric evaluation with the aim of estimating clinical 21
variables (measures) which are generally unavailable in the administrative repositories. 22
As a first step, we verified the database representativeness in evaluating the 23
AHTs-taking behaviour: to this aim, data of the “Fiesole Misurata” study concerning 24
CV diseases, pharmacotherapy and geriatric assessments were compared with those 1
from other epidemiological studies and official statistics. 2
3
METHODS 4
The target population of the “Fiesole Misurata” study database was composed of 5
individuals aged 65 or more living in Fiesole county (Tuscany, Italy). The community 6
living in this area is distributed in nine districts (Fiesole City, Anchetta, Caldine, 7
Compiobbi, Ellera, Girone, Pian del Mugnone, Pian di San Bartolo, San Domenico) and 8
counts 14,264 inhabitants over an area of 42.11 km2 (population density: 340,6 km2). 9
Fiesole citizens have the third highest mean income (€ 17,638 per resident) of Tuscany 10
and the 51st of Italy [15]. 11
Firstly, a list of all residents aged 65 years or more in the community of Fiesole 12
was obtained on May 1st 2010 from the Municipality Registry Office and was merged 13
with the healthcare records obtained from administrative archives of the Local Health 14
Authority was performed by using the citizen’s fiscal code as unique identifier 15
(n=2,228, the claim dataset). Any identification code was automatically converted to a 16
unique anonymous code [16]. 17
Afterwards, all eligible subjects (n=2,228) were contacted by phone, were 18
informed about the study, and were asked for their participation. Three-hundred and 19
eighty-five subjects aged 65 years or more living in the community of Fiesole city 20
decided to participate (n=385, the cross-sectional dataset). Therefore, an appointment 21
was scheduled for each participant and data on multidimensional geriatric assessment 22
(including BP measurement), self-reported drug consumption, and information on 23
socio-demographic status along with lifestyle-related features were collected. 24
The study was approved by the Local Ethic Committee, and all participants 1
signed their informed consent before being interviewed or visited. 2 3 Data collection 4 Claims dataset 5
Admission diagnoses (coded by the International Classification Disease, 9th version, 6
Clinical Modification -ICD9CM) [17-21] and all reimbursed drug prescriptions (coded 7
by the Anatomical Therapeutic Chemical -ATC- classification) were retrospectively 8
obtained for the period between 1 January, 2008 and 31 July , 2010. 9
Hospital admissions (in primary and/or secondary positions) for diabetes 10
(ICD9CM code or antidiabetics use, ATC A10*), ischemic cardiomyopathy, heart 11
failure, haemorrhagic and ischemic stroke, cardiac arrhythmia, were identified. 12
All AHTs pharmacy claims related to Angiotensin-Converting Enzyme (ACE) 13
inhibitors, angiotensin II receptor antagonist (sartans), diuretics, DiHydroPiridine 14
(DHP) Calcium Channel Blockers (CCBs), non-DHP CCBs, beta blockers, peripheral 15
alpha blockers, central inhibitors and the fixed combinations (i.e., ACE inhibitors or 16
sartans or beta blockers with diuretics) were extracted. Furthermore, antithrombotics, 17
antiarrhythmics, lipid lowering drugs and digitalis, as well as the number of ATC 18
categories and hospitalizations being recorded for each elderly resident, were collected. 19
20
Cross-sectional dataset 21
Trained pharmacists interviewed all participants by means of a structured questionnaire 22
on medications use (within the week which preceded the enrolment), socio-23
demographic information (i.e., years of education, marital status) and lifestyle habits 24
(i.e., nutrition, alcohol use and smoking), while six physicians (either geriatricians or 1
clinical pharmacologists) performed the multidimensional assessment and measured the 2
BP. 3
Disability was evaluated with both Instrumental and Basic Activities of Daily 4
Living (IADL and BADL) [22]. Cognitive impairment, depressive or anxiety symptoms 5
were assessed by the Mini Mental State Examination (MMSE) [23] and the Geriatric 6
Depression Scale (GDS) [24]. 7
Blood pressure was measured twice in each arm with the patients in the supine 8
position, after having rested for at least 10 minutes in a quiet room at a comfortable 9
temperature. A cuff larger than the standard was used when arm circumference 10
exceeded 32 cm. The three sets of two BP measures were averaged, and the mean 11
values were considered as the reference systolic and diastolic BP [25]. 12
To evaluate Orthostatic Hypotension (OH), BP was also measured on standing 13
from sitting or supine position according to a time interval of 1, 3 and 5 minutes of 14
standing [26]. 15
Finally, all subjects were required to report previous diagnoses they might have 16
received from a pre-specified list of conditions by answering the question, ‘‘Has your 17
doctor ever told you have…?’’ [27]. All CV diseases being collected by means of 18
claims data were purposely recollected together with asthma, chronic bronchitis, liver 19
diseases, peptic ulcer and cancer [28]. 20
21
Representativeness
22
To verify the representativeness of the “Fiesole Misurata” study database, the following 23
estimates were computed: 24
• prevalence of CV diseases; 1
• prevalence of geriatric-related assessments, based on the standard cut-off points 2
(i.e., BADL ≥1, MMSE≤21, GDS ≥6); 3
• distribution of co-morbidities (i.e., Silver Code scale) [28] and concomitant 4
medications (i.e., count of ATC classes); 5
• prevalence of AHTs use among individuals with self-reported and diagnosed 6
hypertension; 7
• distribution of adherence levels to AHTs. 8
9
Data analysis
10
Percentages, mean values, and related 95% Confidence Intervals (CIs) were computed 11
for categorical and continuous variables,. 12
Proportions of socio-demographic, lifestyle and clinical features (i.e., geriatric 13
assessments, comorbidity and overall medication use) were calculated by using the 14
2,228 residents and 385 survey participants as denominators for claims and cross-15
sectional dataset, respectively. 16
Blood pressure categories were defined by following the official guidelines [9, 17
29-31]. Subjects were diagnosed according to different thresholds, and classified as 18
having ‘Optimal’ (<120/<80 mmHg), ‘Normal’ (120-129/80-84 mmHg), ‘High normal’ 19
(130-139/80-85 mmHg), ‘Hypertension, grade I’ (140-149/90-99 mmHg), 20
‘Hypertension, grade II-III’ (>160/>100 mmHg), ‘Isolate systolic’ (>140/<90 mmHg) 21
BP. The OH was defined as a decrease of at least 20 mm Hg in systolic BP (or systolic 22
BP less than 90 mm Hg) or a decrease of at least 10 mm Hg in diastolic BP when 23
changing from clinostatism to orthostatism [26]. 24
Basic Activities of Daily Living and IADL were registered as continuous and 1
categorical variables. The categorization was obtained by grouping subjects who had 2
lost more than 1 functional autonomy against those who had not lost any of them. 3
According to the literature, MMSE score, which decreases with cognitive impairment, 4
and the GDS score, which increases with depression symptoms, were dichotomized at 5
21 [23] and 6 [24], respectively. The Silver Code was adopted to estimate to the burden 6
of co-morbidity: as per Di Bari and co-workers [28] population was stratified into four 7
prognostic groups based on the individual score (0–3, 4–6, 7–10, and ≥11). 8
With regard to medications, at first, the distribution of AHT classes and other 9
CV medications were computed as proportional values in both claims and cross-10
sectional dataset. Consequently, using the claims data, Drug Daily Dosages 11
(DDDs/1000 inhabitants/day) being prescribed for AHTs as a class and stratified by any 12
single chemical group, were calculated over two years (1 May, 2008- 31 April, 2009 13
versus 1 May, 2009-31 April, 2010). Then, the degree of adherence to AHT was 14
calculated, in claims dataset, among the incident users of AHT. As such, all subjects 15
receiving the first prescription (cohort entry) of AHT from the 1st June 2008 to the 31st 16
February 2010 were identified (i.e., excluding patients prescribed AHTs before the 17
cohort entry). In addition, those with less than 180 days of follow-up after the first 18
prescription were excluded. The adherence was computed as Proportion of Days 19
Covered (PDC), calculated by dividing the cumulative days of AHTs use by the length 20
of follow-up. The number of days supplied from each prescription was calculated by 21
dividing the total amount of active drug in each prescription by the recommended 22
DDDs. All dispensed prescriptions were considered interchangeable. Thus, all overlaps 23
between two or more AHTs prescriptions were subtracted by the total cumulative days 24
of use. When a gap between two treatment periods was ≤90 days, subjects were still 1
considered being on therapy. Therefore, progressively growing adherence was 2
categorized as low with a PDC value <40% , intermediate and high with PDC values 3
40-79% and ≥80%, respectively [10, 32]. According to the subject-specific follow-up, 4
PDC strata were computed at intervals of 6, 12, 18 and 24 months. 5
Finally, subjects who had at least two prescriptions of AHTs, according to their 6
self-reported and diagnosed hypertension, were categorized as ‘self-reported’, ‘mild-7
degree’ (130-139/81-89 mmHg)’ and ‘severe-degree’ (≥140/≥90 mmHg) hypertensive 8 subjects. 9 10 11 RESULTS 12
The claim and cross-sectional dataset consisted of 2,228 and 385 older individuals, 13
respectively. In both datasets, most individuals were females. In the claim dataset, the 14
highest proportion of subjects were less than 70 years, in the cross-sectional dataset the 15
highest proportion of subjects were 70-74 years (Table 1). In the claim dataset, females 16
were older than males, while in the cross-sectional one, age categories were equally 17
distributed between genders. 18
Drugs were purchased in 269 different pharmacies, but three of them covered 19
84% of all dispensed medications. Moreover, patients were assisted by a total of 128 20
general practitioners with eight of them covering 82% of them. 21
In the claim dataset, the burden of comorbidity was lower in females then in 22
males, especially for the highest sub-category of the Silver Code (8.4% versus 15.3%, 23
Table 2). These results were in line with the number of hospitalizations per subject, the 1
number of concomitant medications and the prevalence of hospitalizations due to CV 2
diseases. Among the latters, ischemic cardiomyopathy was 3-fold higher in males than 3
in females, and the corresponding CIs were not overlapped. This picture was maintained 4
among AHTs users, where males outnumbered females for any medication class with 5
the exception of diuretics, central inhibitors and fixed combinations (Table 3). 6
As a whole, the prescribed DDDs were higher in 2009 as compared to 2008 for 7
all AHTs, with the exception of ACE inhibitors (Figure 1). 8
Two-hundred-and-thirty individuals (10.3%of 2,228) constituted the AHT 9
inception dataset. In detail 63.5% were highly adherent to AHTs over the first six 10
months of their treatment, while 14.3% and 22.2% showed intermediate and low levels, 11
respectively (Figure 2). The percentage of the high adherent subjects decreased with 12
time reaching 31.2% at the 24th month. 13
The prevalence of self-reported and diagnosed hypertension was lower in 14
females than in males (Table 4). In contrast, OH was more frequent among females. 15
Subjects who had BP equal to or over than 140/90 mmHg underreported to suffer from 16
hypertension. Specifically, 36/86 (41.9%) females and 28/68 (41.2%) males wrongly 17
reported to be normotensive or mild-hypertensive, respectively (data not shown). With 18
the exception of dyslipidaemia, all CV diseases appeared more common in males, as 19
well as the reduction of cognitive functions (Table 4). On the contrary, females were 20
more functionally impaired and more depressed than men. Taken as whole, disability, 21
cognitive status and depression degree accordingly increased with the participants’ age. 22
The prevalent users of AHTs were slightly higher among females, almost for all 23
medication classes. Only sartans and peripheral alpha blockers were more frequently 24
prescribed in males (Table 5). Diuretics were the most reported medications, followed 1
by ACE inhibitors and sartans (47.0%, 45.3%, and 33.6%, respectively). 2
Generally, almost the 70% of subjects with clinically assessed mild or severe 3
hypertension were pharmacologically treated (Figure 3). 4
5
DISCUSSION 6
This paper describes the methodology with which the representativeness of the “Fiesole 7
Misurata” database was evaluated. To our knowledge, this is the first pharmaco-8
epidemiological tool focused on older subjects which comprises both administrative and 9
clinical information. 10
In the claim dataset, the distribution of age categories was acceptably 11
representative of the Italian older population, although the prevalence of older people 12
was slightly lower than that reported by the official statistics (16% in Fiesole versus 18-13
20% in Italy) [15, 33], and about 25% aged more than 80 years. Concerning the cross-14
sectional dataset, the lower number of younger participants was likely due to self-15
selection of subjects after the proposal of participation.. Indeed, the fact that subjects 16
were instructed about the study topic could have fostered the participation of elders 17
aged more than 70, who knew better their CV conditions and were featured by an higher 18
burden of comorbidity [18, 27, 28, 34]. 19
Also the prevalence of CV diseases was in line with previous results. As shown 20
by “Progetto Cuore” (a comprehensive study on epidemiology of CV diseases in Italy) 21
[8, 35, 36], and in keeping with what was found in other international contexts [3, 5, 6], 22
these diseases are more common in males. On the other hand, the comparison between 23
claim and cross-sectional dataset showed some differences. The fact that acute events 24
(i.e., ischemic cardiomyopathy, stroke, certain arrhythmias) were more frequently 1
reported in the cross-sectional dataset is likely due to the cumulative effect of the self-2
reported diagnoses. In fact, while they can cover the entire life-time period of each 3
participant, the clinical history in claim datasets was limited to the previous two-year 4
period. , Consistently, our cross-sectional estimates agreed with those obtained by Landi 5
and coworkers [18] who enrolled patients with a similar design Also heart failure was 6
more prevalent in the cross-sectional dataset. The discrepancy with claim dataset is 7
likely due to the aforementioned reasons along with the chronic course of this disease 8
[37]. In fact, hospitalizations due to exacerbations of heart failure could occur in a 9
period longer than that we were able to analyse. 10
According to “Fiesole Misurata” study, 27.0% of subjects were classified as 11
functionally impaired. These estimates were in keeping with similar surveys [38, 39]. 12
Accordingly, the prevalence of cognitive status [40], depression [41], OH [26], burden 13
of comorbidities [18, 34] and co-medications [18, 42-45] were consistent with previous 14
estimates. 15
As hypertension was considered, the self-reporting diagnoses underestimated 16
(almost 10% lower) the prevalence of hypertension when compared with the actual BP 17
measurement during the study. Specifically, more than one-third of participants 18
misclassified their BP status; this is in line with the fact that elderly individuals usually 19
underestimate their levels of BP, even if patients’ unawareness of hypertension is 20
recently decreased in western countries [29]. Furthermore, while the percentage of 21
subjects with severe hypertension was higher than 65%, the adherence to AHTs sensibly 22
decreased during the two years after the first prescription. In any case, more than 20% 23
of individuals with severe hypertension did not receive any prescription, and more than 24
30% of the incident users were non-adherent in the first six months of follow-up. These 1
findings demonstrate that the poor AHTs-taking behaviour is quantitatively similar to 2
that reported in the middle-aged population [10, 32]. These results were further 3
strengthened by the fact that the prevalence of each single drug category and the 4
prescribed DDDs agreed with the official prescription reports [46, 47] and previous 5
investigations [48]. 6
From a public health perspective, the “Fiesole Misurata” study could be 7
important in several ways. First of all, it offers a comprehensive picture of a 8
community-based older population in terms of health claim information and clinical 9
features. Furthermore, the quantification of AHTs non-adherence, as well as the 10
measurement of OH, have not been previously reported in an Italian elderly population. 11
Certainly, the present study has limitations. Firstly, the cross-sectional sample 12
has not been randomly selected and it could be therefore affected by selection bias. 13
However, given that all estimates concerning both diseases and medications use were 14
consistent with prior studies, the driven selection of certain patients’ categories should 15
have been minimized. Secondly, some diagnoses coded in claims databases could be 16
underestimated because they are limited to hospital discharge charts. Nevertheless, 17
given that elders are more frequently hospitalized than younger adults, we can assume 18
that underestimation of cardiovascular and other specific diseases (e.g., COPD) is 19
generally negligible in this age category. Finally, claims databases do not comprise the 20
indication of drug use. As a consequence, subjects cannot be differentiated between 21
those who suffer from hypertension and/or heart failure or other conditions. However, 22
the non-adherent behaviour to AHTs equally affects all CV illnesses. 23
Despite these limitations, the present study does not undermine the observed 1
values, particularly considering few Randomized Clinical Trials (RCTs) are conducted 2
in elderly patients, and RCTs often fail to appropriately evaluate the issues related to 3
medications-non-adherence [2]. In particular, differences in drug tolerability, dosing 4
variability, and patient perceptions of the disease are observational (i.e., “real-world”) 5
variables which can remarkably influence the adherence to AHTs. For this reason, 6
appropriate strategies to correct these factors should be implemented. 7
Given that the clinical characteristics of older people residents in Fiesole appear 8
consistent with those of the Italian older population, it is our opinion that further 9
strategies aimed at improving the adherence to AHTs can be implemented and 10
epidemiologically verified by adopting “Fiesole Misurata” study database. 11
ACKNOWLEDGEMENTS 12
This work was conducted with contribution of the Tuscany Region. The authors thank 13
School of Pharmacology and of Geriatrics Specialization for questionnaire 14
administration and data recording, and Fiesole Municipality for data collection. 15
16
CONFLICT OF INTEREST 17
The authors declare that they have no conflict of interest. 18
REFERENCES
1. Cutler DM and Everett W. Thinking outside the pillbox--medication adherence as a priority for health care reform. N Engl J Med 2010; 362(17): 1553-5.
2. Doggrell SA. Adherence to medicines in the older-aged with chronic conditions: does intervention by an allied health professional help? Drugs Aging 2010; 27(3): 239-54. 3. Puddu PE, Menotti A, Tolonen H, Nedeljkovic Sand Kafatos AG. Determinants of
40-year all-cause mortality in the European cohorts of the Seven Countries Study. Eur J Epidemiol 2011; 26(8): 595-608.
4. Konig HH, Heider D, Lehnert T, Riedel-Heller SG, Angermeyer MC, Matschinger H, et al. Health status of the advanced elderly in six European countries: results from a representative survey using EQ-5D and SF-12. Health Qual Life Outcomes 2010; 8: 143.
5. Kim AS and Johnston SC. Global variation in the relative burden of stroke and ischemic heart disease. Circulation 2011; 124(3): 314-23.
6. Kearney PM, Whelton M, Reynolds K, Muntner P, Whelton PKand He J. Global burden of hypertension: analysis of worldwide data. Lancet 2005; 365(9455): 217-23. 7. Wild SH, Fischbacher C, Brock A, Griffiths Cand Bhopal R. Mortality from all causes
and circulatory disease by country of birth in England and Wales 2001-2003. J Public Health (Oxf) 2007; 29(2): 191-8.
8. Palmieri L, Barchielli A, Cesana G, de Campora E, Goldoni CA, Spolaore P, et al. The Italian register of cardiovascular diseases: attack rates and case fatality for
cerebrovascular events. Cerebrovasc Dis 2007; 24(6): 530-9.
9. Mayor S. Hypertension diagnosis should be based on ambulatory blood pressure monitoring, NICE recommends. BMJ 2011; 343: d5421.
10. Mazzaglia G, Ambrosioni E, Alacqua M, Filippi A, Sessa E, Immordino V, et al. Adherence to antihypertensive medications and cardiovascular morbidity among newly diagnosed hypertensive patients. Circulation 2009; 120(16): 1598-605.
11. Choudhry NK, Fischer MA, Avorn J, Liberman JN, Schneeweiss S, Pakes J, et al. The implications of therapeutic complexity on adherence to cardiovascular medications. Arch Intern Med 2011; 171(9): 814-22.
12. Gellad WF, Grenard JLand Marcum ZA. A systematic review of barriers to
medication adherence in the elderly: looking beyond cost and regimen complexity. Am J Geriatr Pharmacother 2011; 9(1): 11-23.
13. Dietlein G and Schroder-Bernhardi D. Use of the mediplus patient database in healthcare research. Int J Clin Pharmacol Ther 2002; 40(3): 130-3.
14. Andersohn F and Garbe E. [Pharmacoepidemiological research with large health databases]. Bundesgesundheitsblatt Gesundheitsforschung Gesundheitsschutz 2008; 51(10): 1135-44.
15. ISTAT. http://www.comuni-italiani.it/048/015/. 2010.
16. Quantin C, Allaert FA, Avillach P, Fassa M, Riandey B, Trouessin G, et al. Building application-related patient identifiers: what solution for a European country? Int J Telemed Appl 2008; 678302.
17. Lloyd-Jones D, Adams R, Carnethon M, De Simone G, Ferguson TB, Flegal K, et al. Heart disease and stroke statistics--2009 update: a report from the American Heart Association Statistics Committee and Stroke Statistics Subcommittee. Circulation 2009; 119(3): 480-6.
18. Landi F, Russo A, Cesari M, Barillaro C, Onder G, Zamboni V, et al. The ilSIRENTE study: a prospective cohort study on persons aged 80 years and older living in a mountain community of Central Italy. Aging Clin Exp Res 2005; 17(6): 486-93. 19. Spolaore P, Brocco S, Fedeli U, Visentin C, Schievano E, Avossa F, et al. Measuring
accuracy of discharge diagnoses for a region-wide surveillance of hospitalized strokes. Stroke 2005; 36(5): 1031-4.
20. Barchielli A, Balzi D, Naldoni P, Roberts AT, Profili F, Dima F, et al. Hospital discharge data for assessing myocardial infarction events and trends, and effects of diagnosis validation according to MONICA and AHA criteria. J Epidemiol
Community Health 2010; doi:10.1136/jech.2010.110908.
21. Tunstall-Pedoe H, Kuulasmaa K, Amouyel P, Arveiler D, Rajakangas AMand Pajak A. Myocardial infarction and coronary deaths in the World Health Organization MONICA Project. Registration procedures, event rates, and case-fatality rates in 38 populations from 21 countries in four continents. Circulation 1994; 90(1): 583-612. 22. Ferrucci L, Guralnik JM, Bandinelli S, Salani B, Del Lungo I, Antonini E, et al., The
development of disability in older persons: relationship with mortality and quality of life., in Ferrucci L, Heikkinen E, Waters WE, Baroni A (editors): Health and quality of life in older Europeans. Florence: INCRAÐ- WHO. 1995. p. 31-66.
23. Folstein MF, Folstein SEand McHugh PR. "Mini-mental state". A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 1975; 12(3): 189-98.
24. Yesavage JA, Brink TL, Rose TL, Lum O, Huang V, Adey M, et al. Development and validation of a geriatric depression screening scale: a preliminary report. J Psychiatr Res 1982; 17(1): 37-49.
25. Di Bari M, Salti F, Nardi M, Pahor M, De Fusco C, Tonon E, et al. Undertreatment of hypertension in community-dwelling older adults: a drug-utilization study in
Dicomano, Italy. J Hypertens 1999; 17(11): 1633-40.
26. Benvenuto LJ and Krakoff LR. Morbidity and mortality of orthostatic hypotension: implications for management of cardiovascular disease. Am J Hypertens 2011; 24(2): 135-44.
27. Di Bari M, Virgillo A, Matteuzzi D, Inzitari M, Mazzaglia G, Pozzi C, et al. Predictive validity of measures of comorbidity in older community dwellers: the Insufficienza Cardiaca negli Anziani Residenti a Dicomano Study. J Am Geriatr Soc 2006; 54(2): 210-6.
28. Di Bari M, Balzi D, Roberts AT, Barchielli A, Fumagalli S, Ungar A, et al. Prognostic stratification of older persons based on simple administrative data: development and validation of the "Silver Code," to be used in emergency department triage. J Gerontol A Biol Sci Med Sci 2010; 65(2): 159-64.
29. Prugger C, Keil U, Wellmann J, de Bacquer D, de Backer G, Ambrosio GB, et al. Blood pressure control and knowledge of target blood pressure in coronary patients across Europe: results from the EUROASPIRE III survey. J Hypertens 2011; 29(8): 1641-1648.
30. Ritchie LD, Campbell NCand Murchie P. New NICE guidelines for hypertension. BMJ 2011; 343: d5644.
31. Scullard P, Abdelhamid A, Steel Nand Qureshi N. Does the evidence referenced in NICE guidelines reflect a primary care population? Br J Gen Pract 2011; 61(584): e112-7.
32. Corrao G, Parodi A, Nicotra F, Zambon A, Merlino L, Cesana G, et al. Better
compliance to antihypertensive medications reduces cardiovascular risk. J Hypertens 2011; 29(3): 610-8.
33. ISTAT. Previsioni della popolazione residente per sesso, età e regione dal 1.1.2001 al 1.1.2051. 2011.
34. Landi F, Liperoti R, Russo A, Capoluongo E, Barillaro C, Pahor M, et al. Disability, more than multimorbidity, was predictive of mortality among older persons aged 80 years and older. J Clin Epidemiol 2010; 63(7): 752-9.
35. Panico S, Palmieri L, Donfrancesco C, Vanuzzo D, Chiodini P, Cesana G, et al. [Reduction of cardiovascular risk and body mass index: findings from the CUORE Project]. G Ital Cardiol (Rome) 2010; 11(5 Suppl 3): 43S-47S.
36. Vanuzzo D, Lo NC, Pilotto L, Palmieri L, Donfrancesco C, Dima F, et al.
[Cardiovascular epidemiologic observatory 2008-2011: preliminary results]. G Ital Cardiol (Rome) 2010; 11(5 Suppl 3): 25S-30S.
37. Kannel WB. Incidence and epidemiology of heart failure. Heart Fail Rev 2000; 5(2): 167-73.
38. Avendano M, Kunst AE, Huisman M, Lenthe FV, Bopp M, Regidor E, et al.
Socioeconomic status and ischaemic heart disease mortality in 10 western European populations during the 1990s. Heart 2006; 92(4): 461-7.
39. Huisani B, Cummings, S., Kilbourne, B., & Roback, H. Group Therapy for Depressed Elderly Women. Group Therapy for Depressed Elderly Women. International Journal of Group Psychotherapy, New York , 54, 3, 295. 2004.
40. EpiCentro. http://www.epicentro.iss.it/focus/anziani/anziani.asp - Il portale dell'epidemiologia per la sanità pubblica. 2010.
41. Friedman O, McAlister FA, Yun L, Campbell NRand Tu K. Antihypertensive drug persistence and compliance among newly treated elderly hypertensives in ontario. Am J Med 2010; 123(2): 173-81.
42. Jyrkka J, Enlund H, Lavikainen P, Sulkava Rand Hartikainen S. Association of polypharmacy with nutritional status, functional ability and cognitive capacity over a three-year period in an elderly population. Pharmacoepidemiol Drug Saf 2011; 20(5): 514-22.
43. Jyrkka J, Mursu J, Enlund Hand Lonnroos E. Polypharmacy and nutritional status in elderly people. Curr Opin Clin Nutr Metab Care 2011;
doi:10.1097/MCO.0b013e32834d155a.
44. Venturini CD, Engroff P, Ely LS, Zago LF, Schroeter G, Gomes I, et al. Gender differences, polypharmacy, and potential pharmacological interactions in the elderly. Clinics (Sao Paulo) 2011; 66(11): 1867-72.
45. Lapi F, Pozzi C, Mazzaglia G, Ungar A, Fumagalli S, Marchionni N, et al.
Epidemiology of suboptimal prescribing in older, community dwellers: a two-wave, population-based survey in Dicomano, Italy. Drugs Aging 2009; 26(12): 1029-38. 46. OsMed. Rapporto sul consumo dei farmaci in Italia. 2009.
47. OsMed. Rapporto sul consumo dei farmaci in Italia. 2010.
48. Poluzzi E, Strahinja P, Vargiu A, Chiabrando G, Silvani MC, Motola D, et al. Initial treatment of hypertension and adherence to therapy in general practice in Italy. Eur J Clin Pharmacol 2005; 61(8): 603-9.
Figure Legends
Figure 1. Use of antihypertensives broken down by the period of use in the claim dataset (DDD/1000 inhabitants/die). ACE: Angiotensin-Converting Enzyme; CCBs : Calcium Channel Blockers; DDDs: Drug Daily Dosages
Figure 2. Degree of adherence among new users of antihypertensives in the AHT dataset. AHT: AntiHyperTensive
Figure 3. Degree of treatment among self-reported and diagnosed hypertensive subjects in the cross-sectional dataset. Mild hypertensive subjects: blood pressure 130-139/81-89 mmHg; Severe hypertensive subjects: blood pressure ≥140/≥90 mmHg; Treated: at least two antihypertensive prescriptions.
Figure 1. Use of antihypertensives broken down by the period of use in the claim dataset (DDD/1000 inhabitants/die). ACE: Angiotensin-Converting Enzyme; CCBs : Calcium Channel Blockers; DDDs: Drug Daily Dosages
Figure 2. Degree of adherence among new users of antihypertensives in the AHT dataset. AHT: AntiHyperTensive
Figure 3. Degree of treatment among self-reported and diagnosed hypertensive subjects in the cross-sectional dataset. Mild hypertensive subjects: blood pressure 130-139/81-89 mmHg; Severe hypertensive subjects: blood pressure ≥140/≥90 mmHg; Treated: at least two antihypertensive prescriptions.
Table 1. Distribution of older subjects’ demographics in the claim (n=2,228) and the cross-sectional (n=385) dataset.
Number Percentage (95% CI)
Overall Females Males
Claims dataset No. of residents 2,228 1,274 954 Age(years) <70 743 33.4 (31.4-35.3) 395 31.0 (28.4-33.6) 348 36.4 (33.4-39.6) 70-74 515 23.1 (21.4-24.9) 295 23.2 (20.9-25.6) 220 23.1 (20.4-25.9) 75-79 413 18.5 (16.9-20.2) 226 17.7 (15.7-19.9) 187 19.6 (17.1-22.2) 80-84 308 13.8 (12.4-15-3) 186 14.6 (12.7-16.6) 122 12.8 (10.7-15.1) >84 249 11.2 (9.9-12.6) 172 13.5 (11.7-15.5) 77 8.1 (6.4-10.0) Cross-sectional dataset No. of participants 385 220 165 Age (years) <70 76 19.7 (15.9-2.1) 41 18.6 (13.7-24.4) 35 21.2 (15.2-28.2) 70-74 92 23.9 (19.1-28.5) 63 28.7 (22.8-35.1) 29 17.6 (12.1-24.3) 75-79 83 21.6 (17.6-26.0) 46 20.9 (15.7-26.9) 37 22.4 (16.3-29.6) 80-84 74 19.2 (15.4-23.5) 39 17.7 (12.9-23.4) 35 21.2 (15.2-28.2) >84 60 15.6 (12.1-19.6) 31 14.1 (9.8-19.4) 29 17.6 (12.1-24.3)
Table 2. Distribution of residents’ clinical features in the claims dataset (n=2,228). Number Percentage (95% CI) Overall (N=2,228) Females (N=1,274) Males (N=954) Silver Code categories
0-3 1,459 65.5 (63.4-67.4) 912 71.6 (68.9-74.0) 547 57.3 (54.1-60.5) 4-6 364 16.3 (14.9-18.0) 138 10.8 (9.2-12.7) 226 23.7 (21.0-26.5) 7-10 152 6.8 (5.8-7.9) 117 9.2 (7.6-10.9) 35 3.7 (2.6-5.1) ≥11 253 11.4 (10.1-12.7) 107 8.4 (6.9-10.0) 146 15.3 (13.1-17.7) Hospitalizations/Subjects a 1,271/2,228 0.6 653/1,274 0.5 618/954 0.6 Number of subjects with
hospital data 663 29.8 (27.9-31.7) 342 26.8 (24.4-29.4) 321 33.6 (30.6-36.4) Prevalent hospitalizations Diabetes
(or antidiabetics: ATC A10*)
313 14.0 (12.6-15.6) 152 11.9 (10.2-13.8) 161 16.7 (14.5-1.94) Ischemic cardiomyopathy 83 3.7 (3.0-4.6) 26 2.0 (1.3-3.0) 57 6.0 (4.6-7.7) Heart failure 67 3.0 (2.3-3.8) 33 2.6 (1.8-3.6) 34 3.6 (2.5-4.9) Haemorrhagic and ischemic
stroke 72 3.2 (2.5-4.0) 38 3.0 (2.1-4.1) 34 3.6 (2.5-4.9) Cardiac Arrhythmia 77 3.5 (2.7-4.3) 37 2.9 (2.1-4.0) 40 4.2 (3.0-5.7) Number of co-prescribed drugs b
mean (±SD) 5.2 (± 5.1) 5.8 (± 4.6) 5.6 (± 5.5) Number of medications 0 1,377 61.8 (59.7-63.8) 824 64.7 (62.0-67.3) 553 58.0 (54.8-61.1) 1-4 578 26.0 (24.1-27.8) 320 25.1 (22.7-27.6) 258 27.0 (24.2-30.0) ≥5 273 12.2 (10.9-13.7) 130 10.2 (8.6-12.0) 143 15.0 (12.8-17.4) a ratio
b any single ATC among medication users
Table 3. Distribution of resident’s use of antihypertensives and other CV medications in the claim dataset (n=2,228). Number Percentage (95% CI) Overall (N=2,228) Females (N=1,274) Males (N=954) Prevalent users of antihypertensives a
Overall 1,507 67.6 (65.6-69.6) 869 68.2 (65.6-70.8) 638 66.9 (63.8-69.8) Age strata (years)
<70 743 54.2 (50.6-57.9) 395 54.2 (49.1-59.2) 348 54.3 (48.9-59.6) 70-74 515 68.3 (64.1-72.3) 295 65.1 (59.3-70.5) 220 72.7 (66.3-78.5) 75-79 413 74.8 (70.3-78.9) 226 76.1 (70.0-81.5) 187 73.3 (66.3-79.5) 80-84 308 76.9 (71.8-81.5) 186 80.6 (74.2-86.1) 122 71.3 (62.4-79.1) >84 249 82.7 (77.4-87.2) 172 82.0 (75.4-87.4) 77 84.4 (74.4-91.7) Medication class a
ACE inhibitors (C09A*) 657 43.6 (41.1-46.1) 352 40.5 (37.2-43.8) 305 47.8 (43.9-51.8) Diuretics (C03*) 536 35.6 (33.1-38.0) 324 37.3 (34.0-40.6) 212 33.2 (29.6-37.0) Sartans (C09C*) 371 24.6 (22.5-26.9) 204 23.5 (20.7-26.4) 167 26.2 (22.8-29.8) Beta blockers (C07A*;
C07EA*) 454 30.1 (27.8-32.5) 248 28.5 (25.5-31.7) 206 32.3 (28.7-36.1) CCBs – DHP (C08CA*) 482 32.0 (29.6-34.4) 271 31.2 (28.1-34.4) 211 33.1 (29.4-36.9) Central inhibitors (C02A*) 26
1.7 (1.1-2.5)
16 1.8 (1.0-3.0)
10 1.6 (0.7-2.9) Alfa blockers, peripheral
(C02C*) 148 9.8 (8.4-11.4) 59 6.8 (5.2-8.7) 89 13.9 (11.3-16.9) CCBs - non DHP (C08CX01; C08D*; C08E*) 100 6.6 (5.4-8.0) 51 5.9 (4.4-7.6) 49 7.7 (5.7-10.0) Beta blockers and diuretics
(C07B*; C07C) 30 2.0 (1.3-2.8) 20 2.3 (1.4-3.5) 10 1.6 (0.7-2.8) ACE inhibitors and
Diuretics (C09B*) 408 27.1 (24.8-29.4) 238 27.4 (24.4-30.5) 170 26.7 (23.2-30.2) Diuretics and Sartans 342
22.7 (20.6-24.9) 210 24.2 (21.3-27.1) 132 20.7 (17.6-24.0) Table 3. continues
Table 3. continued Number Percentage (95% CI) Overall (N=2,228) Females (N=1,274) Males (N=954) Prevalent users of other CV medications
Antithrombotics (B01A*) 1134 50.9 (48.8-53.0) 611 48.0 (45.2-50.7) 523 54.8 (51.7-58.0) Antiarrhythmics (C01B*) 636 28.5 (26.7-30.4) 336 26.4 (24.0-28.8) 300 31.4 (28.5-34.4) Digitalis (C01A*) 131 5.9 (4.9-6.9) 66 5.2 (4.0-6.4) 65 6.8 (5.2-8.4) Lipid lowering (C10*) 540 24.2 (22.5-26.0) 270 21.2 (18.9-23.4) 270 28.3 (25.4-31.2) ACE: Angiotensin-Converting Enzyme
CV: CardioVascular
CCBs : Calcium Channel Blockers DHP: dihydropiridinic
Table 4. Distribution of subject’s clinical features in the cross-sectional dataset (n=385). Number Percentage (95% CI) Overall (N=385) Females (N=220) Males (N=165) BP (mmHg) Optimal: <120/<80 71 18.4 (14.7-22.7) 43 19.5 (14.5-25.4) 28 17.0 (11.6-23.6) Normal: 120-129/80-84 114 29.6 (25.1-34.4) 69 31.4 (25.3-37.9) 45 27.3 (20.6-34.7) High normal: 130-139/80-85 32 8.3 (5.7-11.5) 16 7.3 (4.2-11.5) 16 9.7 (5.6-15.3) Hypertension, grade I: 140-159/90-99 43 11.2 (8.2-14.7) 22 10.0 (6.3-14.7) 21 12.7 (8.0-18.8) Hypertension, grade II-III:
>160/>100 23 6.0 (3.8-8.8) 13 5.9 (3.2-9.9) 10 6.1 (2.9-10.9) Isolate Systolic: >140/<90 88 22.9 (18.8-27.4) 51 23.2 (17.8-29.3) 37 22.4 (16.3-29.6) missing 14 3.6 (0.2-6.0) 6 2.7 (0.1-5.8) 8 4.8 (2.1-9.3) Orthostatic Hypotension a No 306 79.5 (75.1-85.3) 174 79.1 (73.1-84.3) 132 80.0 (73.1-85.8) Yes 48 12.5 (9.3-16.2) 31 14.1 (9.8-19.4) 17 10.3 (6.1-16.0) missing 31 8.0 (5.5-11.2) 15 6.8 (3.8-11.0) 16 9.7 (5.6-15.3) Cardiovascular disease Dyslipidaemia 141 36.6 (31.8-41.6) 96 43.6 (37.0-50.2) 45 27.3 (20.4-34.1) Diabetes (or use of antidiabetic
drugs) 52 13.5 (10.1-16.9) 28 12.7 (8.3-17.2) 24 14.5 (9.1-20.0) Ischemic cardiomyopathy 43 11.2 (8.0-14.3) 20 9.1 (5.3-12.9) 23 13.9 (8.6-19.3) Heart failure 26 6.7 (4.2-9.3) 14 6.4 (3.1-9.6) 12 7.3 (3.3-11.3) Haemorrhagic and ischemic
stroke 27 7.1 (4.4-9.6) 15 6.8 (3.5-10.2) 12 7.3 (3.3-11.3) Self-reported hypertension 222 57.7 (52.7-62.6) 125 56.8 (50.2-63.4) 97 58.8 (51.2-66.4) Table 4. continues
Table 4. continued Number Percentage (95% CI) Overall (N=385) Females (N=220) Males (N=165) Functional status (lost)
BADL, mean (± SD) 0.6 (±1.3) (0.5-0.7) 0.6 (±1.4) (0.5-0.8) 0.5 (±1.3) (0.3-0.7) IADL, mean (± SD) 0.7 (±1.7) (0.5-0.8) 0.8 (±1.8) (0.5-1.0) 0.5 (±1.5) (0.3-0.8) BADL ≥1 Overall 104 27.0 (22.6-31.7) 66 30.0 (24.0-36.5) 38 23.0 (16.8-30.2) Age strata <70 3 4.1 (0.9-11.5) 2 5.3 (0.6-17.7) 1 2.9 (0.7-14.9) 70-74 22 24.2 (15.8-34.3) 15 23.8 (14.0-36.2) 7 25.0 (10.7-44.9) 75-79 26 31.3 (21.6-42.4) 17 37.0 (23.2-52.4) 9 24.3 (11.8-41.2) 80-84 25 34.2 (23.5-46.3) 16 41.0 (25.6-57.9) 9 26.9 (12.9-44.4) >84 28 53.8 (39.5-67.8) 16 57.1 (37.2-75.5) 12 50.0 (29.1-70.9) missing 13 3.4 (1.8-5.7) 6 2.7 (1.0-5.8) 7 4.2 (1.7-8.5) Cognitive status MMSE, mean (± SD) 26.6 (±3.6) (26.3-27.0) 26.7 (±3.6) (26.2-27.2) 26.6 (±3.6) (26.0-27.2) MMSE ≤21 Overall 27 7.0 (4.7-10.0) 11 5.0 (2.5-8.8) 16 9.7 (5.6-15.3) Age strata <70 1 1.3 (0.03-7.3) 1 2.6 (0.07-13.5) - 70-74 2 2.2 (0.3-7.9) 1 1.6 (0.04-8.8) 1 3.6 (0.09-18.3) 75-79 3 3.6 (0.7-10.2) 3 6.5 (1.4-17.9) - 80-84 4 5.4 (1.5-13.3) 1 2.6 (0.07-13.5) 3 8.6 (1.8-23.0) >84 17 30.9 (19.1-44.8) 5 17.9 (6.1-36.9) 12 44.4 (25.5-64.7) missing 10 2.6 (1.2-4.7) 7 3.2 (1.3-6.4) 3 1.8 (0.4-5.2) Table 4. continues
Table 4. continued Number Percentage (95% CI) Overall (N=385) Females (N=220) Males (N=165) Depression GDS, mean (± SD) 3.3 (±2.8) (3.0-3.6) 3.9 (±2.9) (3.5-4.3) 2.5 (±2.5) (2.1-2.9) GDS ≥6 Overall 77 20.0 (16.1-24.3) 54 24.5 (19.0-30.8) 23 13.9 (9.0-20.2) Age strata <70 5 6.8 (2.2-15.1) 4 10.3 (2.9-24.2) 1 2.9 (0.07-14.9) 70-74 21 23.6 (15.2-33.8) 15 24.6 (14.5-37.3) 6 21.4 (8.3-40.9) 75-79 18 21.7 (13.4-32.1) 12 26.1 (14.3-41.1) 6 16.2 (6.2-32.0) 80-84 17 23.3 (14.2-34.6) 12 31.6 (17.5-48.6) 5 14.3 (4.8-30.3) >84 16 30.8 (18.7-45.1) 11 42.3 (23.3-63.1) 5 19.2 (6.5-39.3) missing 14 3.6 (2.0-6.0) 10 4.5(2.2-8.2) 4 2.4 (0.7-6.1) BADL: Basic Activity of Daily Living
BP: Blood Pressure
GDS: Geriatric Depression Scale
IADL: Instrumental Activity of Daily Living MMSE: Mini Mental State Examination SD: standard deviation
a defined as a decrease of at least 20 mm Hg in systolic BP (or systolic BP less than 90 mm Hg) or a decrease of at least 10 mm Hg in diastolic BP when changing from clinostatism to orthostatism.
Table 5. Distribution of subjects’ use of antihypertensives in the cross-sectional dataset (n=385). Number Percentage (95% CI) Overall (N=385) Females (N=220) Males (N=165) Prevalent users of antihypertensives a
Overall 247 64.2 (59.1-68.9) 143 65.0 (58.3-71.3) 104 63.0 (55.2-70.4) Age strata (years)
<70 41 54.0 (42.1-65.4) 23 56.1 (39.7-71.5) 18 51.4 (34.0-68.6) 70-74 55 59.8 (49.0-69.9) 35 55.6 (42.5-68.1) 20 69.0 (49.2-84.7) 75-79 59 71.1 (60.1-80.5) 31 67.4 (52.0-80.5) 28 75.7 (58.8-88.2) 80-84 52 70.3 (58.5-80.3) 28 71.8 (55.1-85.0) 24 68.6 (50.7-83.1) >84 40 66.7 (53.3-78.3) 26 83.9 (66.3-94.5) 14 48.3 (29.4-67.5) Medication class a ACE inhibitors 112 45.3 (39.0-51.8) 66 46.1 (37.8-54.7) 46 44.2 (34.5-54.3) Diuretics 116 47.0 (40.6-53.4) 69 48.2 (39.8-56.7) 47 45.2 (35.4-55.2) Sartans 83 33.6 (27.7-39.9) 46 32.2 (24.6-40.5) 37 35.6 (26.4-45.6) Beta blockers 62 25.1 (19.8-31.0) 43 30.1 (22.7-38.2) 19 18.3 (11.4-27.0) CCBs - DHP 51 20.7 (15.8-26.2) 32 22.4 (15.8-30.1) 19 18.3 (11.4-27.0) Central inhibitors 45 18.2 (13.6-23.6) 33 23.1 (16.4-30.8) 12 11.5 (6.1-19.3) Alfa blockers, peripheral 35
14.2 (10.1-19.1) 8 5.6 (2.4-10.7) 27 26.0 (17.9-25.5) CCBs - non DHP 7 2.8 (1.1-5.7) 5 3.5 (1.1-8.0) 2 1.9 (0.2-6.8) Prevalent users of other CV medications
Antiaggregants 130 33.8 (29.0-38.7) 72 32.7 (26.6-39.4) 58 35.1 (27.9-43.0) Statins 79 20.5 (16.6-24.9) 53 24.1 (10.6-30.3) 26 17.8 (10.6-22-2) ACE: Angiotensin-Converting Enzyme
CV: CardioVascular
CCBs : Calcium Channel Blockers DHP: dihydropiridinic
